Manuka Leaf Chemistry
Melissa Taane
21 February 2019
F1 Layout
Map indicates the layout of the F1 generation resulting from a cross between EC201 and EC103 parents. Column 1 is approximately lengthways facing north.
“Map of F1 Trees”
Height Diagram
Diagram indicates the areas of leaf collection regarding height. Each Tree had 10 leaves collected, 3 from Low, 4 from Mid and 3 from High. The first leaf sampled was measured twice for replication comparison.
“Diagram of leaf collection levels”
Import and Arrange Data
“Full.xlsx” contains measurement information from sampling with the Dualex (https://www.force-a.com/en/capteurs-optiques-optical-sensors/dualex-scientific-chlorophyll-meter/), including;
Surface content of chlorophyll in µg/cm² (Chl)
The index of epidermal flavanols (Flav)
Nitrogen Balance Index status (NBI)
Epidermal Anthocyanins (Anth).
It also contains information about the block position, the leaf height information, and presense or absence of flowering
Sheet “Dup” contains only the samples that were replicated.
# Import Data Measures
Data <-read.xlsx("Full.xlsx", sheetName ="Full")
head(Data)## Collection.Day Allocation Block Column Row group Group.ID Tree.ID Rep.
## 1 4 CG 0 0 0 20 OG CG N
## 2 4 CG 0 0 0 20 OG CG N
## 3 4 CG 0 0 0 20 OG CG N
## 4 4 CG 0 0 0 20 OG CG N
## 5 4 CG 0 0 0 20 OG CG N
## 6 4 CG 0 0 0 20 OG CG N
## measure Height Flower Chl Flav Anth NBI
## 1 9 H Y 28.220 2.418 0.205 11.67
## 2 3 L Y 27.958 2.298 0.691 12.17
## 3 11 H Y 33.727 2.458 0.527 13.72
## 4 4 L Y 25.938 1.758 0.172 14.76
## 5 10 H Y 36.205 2.283 0.270 15.86
## 6 6 M Y 34.332 2.115 0.150 16.23# Import Replicate Data
Dup <-read.xlsx("Full.xlsx", sheetName ="Dup")
head(Dup)## Collection.Day group Group.ID Tree.ID Rep. measure Height Chl Flav
## 1 2.0 23 60 IN4DV Y1 1 L 1.916 2.363
## 2 1.5 3 1 IN4BT Y1 1 L 3.124 2.300
## 3 2.0 17 54 IN4DL Y2 2 L 3.414 1.826
## 4 1.5 32 28 IN4CP Y2 2 L 4.097 1.943
## 5 2.0 9 46 IN4DC Y1 1 L 4.909 1.928
## 6 1.5 5 3 IN4BW Y1 1 L 4.924 1.848
## Anth NBI
## 1 0.174 0.81
## 2 0.191 1.36
## 3 0.112 1.87
## 4 0.051 2.11
## 5 0.103 2.55
## 6 0.165 2.66#Isolate Crimson Glory Outgroup
CG = Data[c(1:11),]
head(CG)## Collection.Day Allocation Block Column Row group Group.ID Tree.ID Rep.
## 1 4 CG 0 0 0 20 OG CG N
## 2 4 CG 0 0 0 20 OG CG N
## 3 4 CG 0 0 0 20 OG CG N
## 4 4 CG 0 0 0 20 OG CG N
## 5 4 CG 0 0 0 20 OG CG N
## 6 4 CG 0 0 0 20 OG CG N
## measure Height Flower Chl Flav Anth NBI
## 1 9 H Y 28.220 2.418 0.205 11.67
## 2 3 L Y 27.958 2.298 0.691 12.17
## 3 11 H Y 33.727 2.458 0.527 13.72
## 4 4 L Y 25.938 1.758 0.172 14.76
## 5 10 H Y 36.205 2.283 0.270 15.86
## 6 6 M Y 34.332 2.115 0.150 16.23#Isolate East Cape 201 Parent
EC201 = Data[c(12:21),]
head(EC201)## Collection.Day Allocation Block Column Row group Group.ID Tree.ID Rep.
## 12 4 EC201 0 0 0 18 P1 EC201 N
## 13 4 EC201 0 0 0 18 P1 EC201 N
## 14 4 EC201 0 0 0 18 P1 EC201 N
## 15 4 EC201 0 0 0 18 P1 EC201 N
## 16 4 EC201 0 0 0 18 P1 EC201 N
## 17 4 EC201 0 0 0 18 P1 EC201 N
## measure Height Flower Chl Flav Anth NBI
## 12 9 H Y 30.722 2.138 0.156 14.37
## 13 6 M Y 45.272 1.861 0.075 24.33
## 14 10 H Y 46.174 1.868 0.052 24.72
## 15 8 M Y 57.480 1.897 0.037 30.30
## 16 5 M Y 57.752 1.681 0.125 34.35
## 17 3 L Y 54.927 1.592 0.061 34.50#Isolate East Cape 103 Parent
EC103 = Data[c(22:33),]
head(EC103)## Collection.Day Allocation Block Column Row group Group.ID Tree.ID Rep.
## 22 4 EC103 0 0 0 19 P2 EC103 N
## 23 4 EC103 0 0 0 19 P2 EC103 N
## 24 4 EC103 0 0 0 19 P2 EC103 N
## 25 4 EC103 0 0 0 19 P2 EC103 N
## 26 4 EC103 0 0 0 19 P2 EC103 N
## 27 4 EC103 0 0 0 19 P2 EC103 N
## measure Height Flower Chl Flav Anth NBI
## 22 8 M Y 31.471 1.324 0.018 23.77
## 23 11 H Y 57.349 2.111 0.053 27.17
## 24 4 L Y 43.949 1.478 0.004 29.73
## 25 12 H Y 56.779 1.846 0.108 30.75
## 26 9 M Y 55.916 1.746 0.055 32.03
## 27 10 H Y 57.809 1.746 0.054 33.11#Isolate Offspring from the Parental Cross
F1 = Data[c(34:1825),]
head(F1)## Collection.Day Allocation Block Column Row group Group.ID Tree.ID Rep.
## 34 1 F1 1 1 1 15 1 IN4BT N
## 35 1 F1 1 1 1 15 1 IN4BT N
## 36 1 F1 1 1 1 15 1 IN4BT N
## 37 1 F1 1 1 1 15 1 IN4BT N
## 38 1 F1 1 1 1 15 1 IN4BT N
## 39 1 F1 1 1 1 15 1 IN4BT Y1
## measure Height Flower Chl Flav Anth NBI
## 34 8 M N 3.268 1.958 0.143 1.67
## 35 2 L N 23.326 2.581 0.065 9.04
## 36 11 M N 24.682 2.265 0.112 10.90
## 37 7 M N 27.068 2.229 0.084 12.14
## 38 14 H N 27.459 2.135 0.138 12.86
## 39 3 L N 34.592 2.583 0.048 13.39Checking Similarity of Replicates
Replicates were taken by measuring a leaf sample of each tree twice, in order to establish consistency and reliability of measurements with the Dualex.
R <-group_by(Data,Rep.)Replicate Means
#Chlorophyll
RC = mean(R$Chl);RC## [1] 35.21443#Nitrogen Index
RN = mean(R$NBI);RN## [1] 18.34807#Anthocyanin
RA = mean(R$Anth);RA## [1] 0.09772877#Flavanoids
RF = mean(R$Flav);RF## [1] 1.964405Replicate Plots
plot<- ggplot(Dup) + aes(x = Rep., y = Chl) + geom_boxplot()+geom_hline(yintercept = RC)
plot2<- ggplot(Dup) + aes(x = Rep., y = NBI ) + geom_boxplot()+geom_hline(yintercept = RN)
plot3<- ggplot(Dup) + aes(x = Rep., y = Anth ) + geom_boxplot()+geom_hline(yintercept = RA)
plot4<- ggplot(Dup) + aes(x = Rep., y = Flav ) + geom_boxplot()+geom_hline(yintercept = RF)
grid.arrange(plot,plot2,plot3,plot4, ncol = 4)
ANOVA of Replicates
#Chlorophyll
mod <-lm(Dup$Chl ~ Dup$Rep.)
anova(mod)## Analysis of Variance Table
##
## Response: Dup$Chl
## Df Sum Sq Mean Sq F value Pr(>F)
## Dup$Rep. 1 218 218.33 1.033 0.3102
## Residuals 314 66366 211.36#Flavanol
mod2 <-lm(Dup$Flav ~ Dup$Rep.)
anova(mod2)## Analysis of Variance Table
##
## Response: Dup$Flav
## Df Sum Sq Mean Sq F value Pr(>F)
## Dup$Rep. 1 0.0521 0.052129 0.6506 0.4205
## Residuals 314 25.1596 0.080126#Anthocyanin
mod3 <-lm(Dup$Anth ~ Dup$Rep.)
anova(mod3)## Analysis of Variance Table
##
## Response: Dup$Anth
## Df Sum Sq Mean Sq F value Pr(>F)
## Dup$Rep. 1 0.00003 0.0000304 0.0151 0.9023
## Residuals 314 0.63303 0.0020160#Nitrogen
mod4 <-lm(Dup$NBI ~ Dup$Rep.)
anova(mod4)## Analysis of Variance Table
##
## Response: Dup$NBI
## Df Sum Sq Mean Sq F value Pr(>F)
## Dup$Rep. 1 82.5 82.530 1.115 0.2918
## Residuals 314 23241.0 74.016The absence of statistically significant results indicates that our replicates are likely to be consistent.
Comparing Allocations
Allocation refers to which group measurements were taken from, i.e. A Parental Tree (EC103, EC201), Outgroup Tree (CG), F1 Cross offspring
AL <-group_by(Data,Allocation)Measurement Means
#Chlorophyll
AlC = mean(AL$Chl);AlC## [1] 35.21443#Nitrogen
AlN = mean(AL$NBI);AlN## [1] 18.34807#Flavanoids
AlF = mean(AL$Flav);AlF## [1] 1.964405#Anthocyanin
AlA = mean(AL$Anth);AlA## [1] 0.09772877Boxplot comparing medians and measurement distributions
plot5<- ggplot(Data) + aes(x = Allocation, y = Chl) + geom_boxplot()+geom_hline(yintercept = AlC)
plot6<- ggplot(Data) + aes(x = Allocation, y = NBI) + geom_boxplot()+geom_hline(yintercept = AlN)
plot7<- ggplot(Data) + aes(x = Allocation, y = Flav) + geom_boxplot()+geom_hline(yintercept = AlF)
plot8<- ggplot(Data) + aes(x = Allocation, y = Anth) + geom_boxplot()+geom_hline(yintercept = AlA)
grid.arrange(plot5,plot6,plot7,plot8, ncol=2)
ANOVAs
#Chlorophyll
mod5 <-lm(Data$Chl ~ Data$Allocation)
anova(mod5)## Analysis of Variance Table
##
## Response: Data$Chl
## Df Sum Sq Mean Sq F value Pr(>F)
## Data$Allocation 3 6742 2247.21 9.9149 1.748e-06 ***
## Residuals 1821 412728 226.65
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1#Nitrogen
mod6 <-lm(Data$NBI ~ Data$Allocation)
anova(mod5)## Analysis of Variance Table
##
## Response: Data$Chl
## Df Sum Sq Mean Sq F value Pr(>F)
## Data$Allocation 3 6742 2247.21 9.9149 1.748e-06 ***
## Residuals 1821 412728 226.65
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1#Flavanol
mod7 <-lm(Data$Flav ~ Data$Allocation)
anova(mod7)## Analysis of Variance Table
##
## Response: Data$Flav
## Df Sum Sq Mean Sq F value Pr(>F)
## Data$Allocation 3 2.713 0.90423 14.755 1.708e-09 ***
## Residuals 1821 111.598 0.06128
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1#Anthocyanin
mod8 <-lm(Data$Anth ~ Data$Allocation)
anova(mod8)## Analysis of Variance Table
##
## Response: Data$Anth
## Df Sum Sq Mean Sq F value Pr(>F)
## Data$Allocation 3 0.3040 0.101343 47.474 < 2.2e-16 ***
## Residuals 1821 3.8873 0.002135
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1Parental Cross Analysis (F1 Generation)
Investigating measures across the F1 cross population
O <-group_by(F1,Tree.ID)
SO <-summarise(O,count = n(), Chl = mean(Chl,na.rm =T), NBI = mean(NBI, na.rm=T), Anth = mean(Anth, ra.rm = T), Flav = mean(Flav, na.rm = T));SO## # A tibble: 159 x 6
## Tree.ID count Chl NBI Anth Flav
## <fct> <int> <dbl> <dbl> <dbl> <dbl>
## 1 IN4BT 26 34.5 16.2 0.110 2.17
## 2 IN4BV 11 27.5 13.8 0.121 2.01
## 3 IN4BW 11 24.8 12.5 0.127 1.99
## 4 IN4BX 11 37.6 18.4 0.0905 2.05
## 5 IN4BY 10 24.2 12.5 0.110 1.98
## 6 IN4BZ 11 32.0 15.6 0.106 2.09
## 7 IN4C0 11 28.0 13.4 0.139 2.04
## 8 IN4C1 11 30.8 15.1 0.118 2.04
## 9 IN4C2 11 42.6 21.2 0.0878 2.04
## 10 IN4C3 11 25.7 14.1 0.106 1.97
## # ... with 149 more rowsSurface Content of Chlorophyll (µg/cm²)
#Mean
OC <- mean(SO$Chl);OC## [1] 35.13564#Individual Tree Measures
plot15<-ggplot(O) + aes(x = Tree.ID, y = Chl, group_by(Tree.ID)) + geom_boxplot() +geom_hline(yintercept = OC)
#Measure Distribution
plot16<-ggplot(O) + aes(x = Chl) + geom_histogram()
grid.arrange(plot15,plot16)## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
#ANOVA
ModChl <- lm(O$Chl ~ O$Tree.ID)
anova(ModChl)## Analysis of Variance Table
##
## Response: O$Chl
## Df Sum Sq Mean Sq F value Pr(>F)
## O$Tree.ID 158 47130 298.29 1.3395 0.004474 **
## Residuals 1633 363634 222.68
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1Index of Epidermal Flavnols
OF <- mean(SO$Flav);OF## [1] 1.965265plot17<-ggplot(O) + aes(x = Tree.ID, y = Flav) + geom_boxplot() +geom_hline(yintercept = OF)
plot18 <-ggplot(O) + aes(x = Flav) + geom_histogram()
grid.arrange(plot17,plot18)## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
ModFlav <- lm(O$Flav ~ O$Tree.ID)
anova(ModFlav)## Analysis of Variance Table
##
## Response: O$Flav
## Df Sum Sq Mean Sq F value Pr(>F)
## O$Tree.ID 158 25.118 0.15897 3.0531 < 2.2e-16 ***
## Residuals 1633 85.030 0.05207
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1Epidermal Anthocyanin
OA <- mean(SO$Anth);OA## [1] 0.09699261plot19<-ggplot(O) + aes(x = Tree.ID, y = Anth) + geom_boxplot() +geom_hline(yintercept = OA)
plot20<-ggplot(O) + aes(x = Anth) + geom_histogram()
grid.arrange(plot19,plot20)## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
ModAnth <- lm(O$Anth ~ O$Tree.ID)
anova(ModAnth)## Analysis of Variance Table
##
## Response: O$Anth
## Df Sum Sq Mean Sq F value Pr(>F)
## O$Tree.ID 158 0.5253 0.0033247 1.8093 2.001e-08 ***
## Residuals 1633 3.0007 0.0018376
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1Nitrogen Balance Index Status
#Mean
ON <- mean(SO$NBI);ON## [1] 18.27171plot21<-ggplot(O) + aes(x = Tree.ID, y = NBI) + geom_boxplot() +geom_hline(yintercept = ON)
plot22<-ggplot(O) + aes(x = NBI) + geom_histogram()
grid.arrange(plot21,plot22)## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
ModNBI <- lm(O$NBI ~ O$Tree.ID)
anova(ModNBI)## Analysis of Variance Table
##
## Response: O$NBI
## Df Sum Sq Mean Sq F value Pr(>F)
## O$Tree.ID 158 16378 103.661 1.5147 8.435e-05 ***
## Residuals 1633 111761 68.439
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1Comparing Measures from the 3 height categories
Comparisons of measures between Low, Mid and High locations on trees.
Hgt <- group_by(F1, Height)
SHgt=summarise(Hgt, count = n(), Chl = mean(Chl,na.rm=T), NBI = mean(NBI, na.rm = T), Anth = mean(Anth, na.rm = T), Flav = mean(Flav, na.rm = T))
#Overall means across all heights
#Chlorophyll
HC = mean(SHgt$Chl);HC## [1] 34.91248#Nitrogen
HN = mean(SHgt$NBI);HN## [1] 18.15982#Anthocyanin
HA = mean(SHgt$Anth);HA## [1] 0.09770524#Flavanoids
HF = mean(SHgt$Flav);HF## [1] 1.96426#Boxplots visualising height comparisons
plot23<-ggplot(Hgt) + aes(x = Height, y = NBI) + geom_boxplot()+geom_hline(yintercept = HN)
plot24<-ggplot(Hgt) + aes(x = Height, y = Chl) + geom_boxplot()+geom_hline(yintercept = HC)
plot25<-ggplot(Hgt) + aes(x = Height, y = Flav) + geom_boxplot()+geom_hline(yintercept = HF)
plot26<-ggplot(Hgt) + aes(x = Height, y = Anth) + geom_boxplot()+geom_hline(yintercept = HA)
grid.arrange(plot23,plot24,plot25,plot26,ncol=2)
#Visualising Measurement Distributions
plot231<-ggplot(Hgt) + aes(x = Chl) + geom_histogram()
plot241<-ggplot(Hgt) + aes(x = NBI) + geom_histogram()
plot251<-ggplot(Hgt) + aes(x = Flav) + geom_histogram()
plot261<-ggplot(Hgt) + aes(x = Anth) + geom_histogram()
grid.arrange(plot231,plot241,plot251,plot261)## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
#ANOVA
#Chlorophyll
FitH <- lm(Hgt$Chl ~ Hgt$Height)
anova(FitH)## Analysis of Variance Table
##
## Response: Hgt$Chl
## Df Sum Sq Mean Sq F value Pr(>F)
## Hgt$Height 2 853 426.30 1.8605 0.1559
## Residuals 1789 409911 229.13#Nitrogen
FitH2 <- lm(Hgt$NBI ~ Hgt$Height)
anova(FitH2)## Analysis of Variance Table
##
## Response: Hgt$NBI
## Df Sum Sq Mean Sq F value Pr(>F)
## Hgt$Height 2 100 50.044 0.6992 0.4971
## Residuals 1789 128039 71.570#Flavanoids
FitH3 <- lm(Hgt$Flav ~ Hgt$Height)
anova(FitH3)## Analysis of Variance Table
##
## Response: Hgt$Flav
## Df Sum Sq Mean Sq F value Pr(>F)
## Hgt$Height 2 0.915 0.45758 7.4943 0.0005739 ***
## Residuals 1789 109.232 0.06106
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1#Anthocyanin
FitH4 <- lm(Hgt$Anth ~ Hgt$Height)
anova(FitH4)## Analysis of Variance Table
##
## Response: Hgt$Anth
## Df Sum Sq Mean Sq F value Pr(>F)
## Hgt$Height 2 0.0526 0.0262759 13.533 1.468e-06 ***
## Residuals 1789 3.4735 0.0019416
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1Correlations Between Measures
Comparing correlations between the 4 dualex measures
cor(Data[c(13:16)])## Chl Flav Anth NBI
## Chl 1.0000000 -0.1060143 -0.3038865 0.9468570
## Flav -0.1060143 1.0000000 0.1845141 -0.3865972
## Anth -0.3038865 0.1845141 1.0000000 -0.3204378
## NBI 0.9468570 -0.3865972 -0.3204378 1.0000000plot9 <-ggplot(F1) + aes(x = Chl, y = Anth) + geom_point()
plot10 <-ggplot(F1) + aes(x = NBI, y = Anth) + geom_point()
plot11<-ggplot(F1) + aes(x = Flav, y = Anth) + geom_point()
plot12<-ggplot(F1) + aes(x = NBI, y = Chl) + geom_point()
plot13<-ggplot(F1) + aes(x = Flav, y = Chl) + geom_point()
plot14<-ggplot(F1) + aes(x = Flav, y = NBI) + geom_point()
grid.arrange(plot9,plot10,plot11,plot12,plot13,plot14)
comp <-lm(Flav ~ Anth, data = F1)
anova(comp)## Analysis of Variance Table
##
## Response: Flav
## Df Sum Sq Mean Sq F value Pr(>F)
## Anth 1 2.897 2.89687 48.348 4.983e-12 ***
## Residuals 1790 107.251 0.05992
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1comp2 <-lm(NBI ~ Anth, data = F1)
anova(comp2)## Analysis of Variance Table
##
## Response: NBI
## Df Sum Sq Mean Sq F value Pr(>F)
## Anth 1 14102 14101.7 221.35 < 2.2e-16 ***
## Residuals 1790 114037 63.7
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1comp3 <-lm(Flav ~ NBI, data = F1)
anova(comp3)## Analysis of Variance Table
##
## Response: Flav
## Df Sum Sq Mean Sq F value Pr(>F)
## NBI 1 14.948 14.9476 281.05 < 2.2e-16 ***
## Residuals 1790 95.200 0.0532
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1comp4 <-lm(Flav ~ Anth*NBI, data = F1)
anova(comp4)## Analysis of Variance Table
##
## Response: Flav
## Df Sum Sq Mean Sq F value Pr(>F)
## Anth 1 2.897 2.8969 54.5333 2.332e-13 ***
## NBI 1 12.248 12.2485 230.5764 < 2.2e-16 ***
## Anth:NBI 1 0.022 0.0217 0.4084 0.5229
## Residuals 1788 94.980 0.0531
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1comp5 <-lm(Anth ~ Flav*NBI, data = F1)
anova(comp5)## Analysis of Variance Table
##
## Response: Anth
## Df Sum Sq Mean Sq F value Pr(>F)
## Flav 1 0.09273 0.092735 53.966 3.087e-13 ***
## NBI 1 0.30182 0.301822 175.641 < 2.2e-16 ***
## Flav:NBI 1 0.05898 0.058981 34.323 5.544e-09 ***
## Residuals 1788 3.07250 0.001718
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1comp6 <-lm(NBI ~ Flav*Anth, data = F1)
anova(comp6)## Analysis of Variance Table
##
## Response: NBI
## Df Sum Sq Mean Sq F value Pr(>F)
## Flav 1 17389 17389.2 311.087 < 2.2e-16 ***
## Anth 1 9736 9736.0 174.175 < 2.2e-16 ***
## Flav:Anth 1 1068 1068.0 19.106 1.307e-05 ***
## Residuals 1788 99946 55.9
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1Column Block Effects
Cmn <- group_by(F1, Column)
SCmn <- summarise(Cmn, count =n(),Chl = mean(Chl,na.rm=T), NBI = mean(NBI, ra.nm = T), Anth = mean(Anth, na.rm=T),Flav = mean(Flav,na.rm =T))
CA = mean(SCmn$Anth);CA## [1] 0.09596523CC = mean(SCmn$Chl);CC## [1] 34.87978CN = mean(SCmn$NBI);CN## [1] 18.30061CF = mean(SCmn$Flav);CF## [1] 1.947604plot27<-ggplot(F1) + aes(x = Column, y = Anth,group = Column) + geom_boxplot()+geom_hline(yintercept = CA)
plot28<-ggplot(F1) + aes(x = Column, y = Flav,group = Column) + geom_boxplot() +geom_hline(yintercept = CF)
plot29<-ggplot(F1) + aes(x = Column, y = NBI,group = Column) + geom_boxplot() +geom_hline(yintercept = CN)
plot30<-ggplot(F1) + aes(x = Column, y = Chl,group = Column) + geom_boxplot() +geom_hline(yintercept = CC)
grid.arrange(plot27,plot28,plot29,plot30)
Cmn$Column = as.factor(Cmn$Column)
Col <- lm(Cmn$Anth ~ Cmn$Column)
anova(Col)## Analysis of Variance Table
##
## Response: Cmn$Anth
## Df Sum Sq Mean Sq F value Pr(>F)
## Cmn$Column 3 0.0122 0.0040655 2.0687 0.1024
## Residuals 1788 3.5138 0.0019652Col2 <- lm(Cmn$Flav ~ Cmn$Column)
anova(Col2)## Analysis of Variance Table
##
## Response: Cmn$Flav
## Df Sum Sq Mean Sq F value Pr(>F)
## Cmn$Column 3 2.305 0.76823 12.737 3.087e-08 ***
## Residuals 1788 107.843 0.06031
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1Col3 <- lm(Cmn$NBI ~ Cmn$Column)
anova(Col3)## Analysis of Variance Table
##
## Response: Cmn$NBI
## Df Sum Sq Mean Sq F value Pr(>F)
## Cmn$Column 3 821 273.626 3.8427 0.009333 **
## Residuals 1788 127318 71.207
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1Col4 <- lm(Cmn$Chl ~ Cmn$Column)
anova(Col4)## Analysis of Variance Table
##
## Response: Cmn$Chl
## Df Sum Sq Mean Sq F value Pr(>F)
## Cmn$Column 3 1586 528.66 2.3101 0.07453 .
## Residuals 1788 409178 228.85
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1Row Block Effects
Rw <-group_by(F1, Row)
SRw =summarise(Rw, count = n(), Chl = mean(Chl, na.rm=T),NBI = mean(NBI, na.rm=T), Flav = mean(Flav, na.rm = T), Anth = mean(Anth, na.rm=T))
#Means
#Chlorophyll
RwC = mean(SRw$Chl);RwC## [1] 34.93729#Anthocyanin
RwA = mean(SRw$Anth);RwA## [1] 0.09720837#Flavanoid
RwF = mean(SRw$Flav);RwF## [1] 1.965176#Nitrogen
RwN = mean(SRw$NBI);RwN## [1] 18.17205#Comparing Tree Measures
plot31<-ggplot(Rw) + aes(x = Row, y = Chl, group = Row) + geom_boxplot() +geom_hline(yintercept = RwC);plot31
plot32<-ggplot(Rw) + aes(x = Row, y = Anth, group = Row) + geom_boxplot() +geom_hline(yintercept = RwA);plot32
plot33<-ggplot(Rw) + aes(x = Row, y = Flav,group = Row) + geom_boxplot() +geom_hline(yintercept = RwF);plot33
plot34<-ggplot(Rw) + aes(x = Row, y = NBI, group = Row) + geom_boxplot() +geom_hline(yintercept = RwN);plot34
#ANOVA
#Anthocyanin
Ro <- lm(Rw$Anth ~ Rw$Row)
anova(Ro)## Analysis of Variance Table
##
## Response: Rw$Anth
## Df Sum Sq Mean Sq F value Pr(>F)
## Rw$Row 49 0.2346 0.0047878 2.5339 4.261e-08 ***
## Residuals 1742 3.2914 0.0018895
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1#Nitrogen
Ro2 <- lm(Rw$NBI ~ Rw$Row)
anova(Ro2)## Analysis of Variance Table
##
## Response: Rw$NBI
## Df Sum Sq Mean Sq F value Pr(>F)
## Rw$Row 49 7866 160.527 2.325 8.14e-07 ***
## Residuals 1742 120273 69.043
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1#Flavanoids
Ro3 <- lm(Rw$Flav ~ Rw$Row)
anova(Ro3)## Analysis of Variance Table
##
## Response: Rw$Flav
## Df Sum Sq Mean Sq F value Pr(>F)
## Rw$Row 49 8.215 0.167646 2.865 3.11e-10 ***
## Residuals 1742 101.933 0.058515
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1#Chlorophyll
Ro4 <- lm(Rw$Chl ~ Rw$Row)
anova(Ro4)## Analysis of Variance Table
##
## Response: Rw$Chl
## Df Sum Sq Mean Sq F value Pr(>F)
## Rw$Row 49 21394 436.61 1.9533 0.0001046 ***
## Residuals 1742 389370 223.52
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1Block Effects, Combining Row and Columns.
Z = group_by(F1, Tree.ID,Row,Column)
Z1<- summarise(Z, count = n(), Chl = mean(Chl, na.rm=T),NBI = mean(NBI, na.rm=T), Flav = mean(Flav, na.rm = T), Anth = mean(Anth, na.rm=T))ZChl = Z1[c(2,3,5)] MChl = mean(ZChl\(Chl);MChl ZChl\)Mean = ZChl$Chl-MChl ZChl = ZChl[-c(3)] write.xlsx(as.data.frame(ZChl), file = “Chl.xlsx”, sheetName =“Chl”,col.names = T, row.names = F)
ZNBI = Z1[c(2,3,6)] MNBI = mean(ZNBI\(NBI);MNBI ZNBI\)Mean = ZNBI$NBI-MNBI ZNBI = ZNBI[-c(3)] write.xlsx(as.data.frame(ZNBI), file = “NBI.xlsx”, sheetName =“NBI”,col.names = T, row.names = F)
ZFlav = Z1[c(2,3,7)] MFlav = mean(ZFlav\(Flav);MFlav ZFlav\)Mean = ZFlav$Flav-MFlav ZFlav = ZFlav[-c(3)] write.xlsx(as.data.frame(ZFlav), file = “Flav.xlsx”, sheetName =“Flav”,col.names = T, row.names = F)
ZAnth = Z1[c(2,3,8)] MAnth = mean(ZAnth\(Anth);MAnth ZAnth\)Mean = ZAnth$Anth-MAnth ZAnth = ZAnth[-c(3)] write.xlsx(as.data.frame(ZAnth), file = “Anth.xlsx”, sheetName =“Anth”,col.names = T, row.names = F)
F1$Column = as.factor(F1$Column)
AnthDoc <-read.xlsx("Anth.xlsx", sheetName = "Anth")
heatmaply(AnthDoc, xlab = "Column",ylab ="Row", Rowv = FALSE, Colv = FALSE)ARC <-lm(F1$Anth ~ F1$Row + F1$Column)
anova(ARC)## Analysis of Variance Table
##
## Response: F1$Anth
## Df Sum Sq Mean Sq F value Pr(>F)
## F1$Row 49 0.2346 0.0047878 2.5369 4.088e-08 ***
## F1$Column 3 0.0095 0.0031656 1.6774 0.1699
## Residuals 1739 3.2819 0.0018873
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1ARC2 <- lm(F1$Anth ~ F1$Row*F1$Column)
anova(ARC2)## Analysis of Variance Table
##
## Response: F1$Anth
## Df Sum Sq Mean Sq F value Pr(>F)
## F1$Row 49 0.2346 0.0047878 2.6055 1.593e-08 ***
## F1$Column 3 0.0095 0.0031656 1.7227 0.16034
## F1$Row:F1$Column 106 0.2812 0.0026529 1.4437 0.00276 **
## Residuals 1633 3.0007 0.0018376
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1ChlDoc<-read.xlsx("Chl.xlsx", sheetName = "Chl")
heatmaply(ChlDoc, xlab = "Column",ylab ="Row", Rowv = FALSE, Colv = FALSE)CRC <-lm(F1$Chl ~ F1$Row + F1$Column)
anova(CRC)## Analysis of Variance Table
##
## Response: F1$Chl
## Df Sum Sq Mean Sq F value Pr(>F)
## F1$Row 49 21394 436.61 1.9606 9.568e-05 ***
## F1$Column 3 2118 706.12 3.1709 0.02344 *
## Residuals 1739 387252 222.69
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1CRC2 <- lm(F1$Chl ~ F1$Row*F1$Column)
anova(CRC2)## Analysis of Variance Table
##
## Response: F1$Chl
## Df Sum Sq Mean Sq F value Pr(>F)
## F1$Row 49 21394 436.61 1.9607 9.766e-05 ***
## F1$Column 3 2118 706.12 3.1710 0.02346 *
## F1$Row:F1$Column 106 23617 222.81 1.0006 0.48183
## Residuals 1633 363634 222.68
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1NBIDoc<-read.xlsx("NBI.xlsx", sheetName = "NBI")
heatmaply(NBIDoc, xlab = "Column",ylab ="Row", Rowv = FALSE, Colv = FALSE)NRC <-lm(F1$NBI ~ F1$Row + F1$Column)
anova(NRC)## Analysis of Variance Table
##
## Response: F1$NBI
## Df Sum Sq Mean Sq F value Pr(>F)
## F1$Row 49 7866 160.527 2.3364 6.969e-07 ***
## F1$Column 3 789 263.021 3.8281 0.009527 **
## Residuals 1739 119484 68.708
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1NRC2 <- lm(F1$NBI ~ F1$Row*F1$Column)
anova(NRC2)## Analysis of Variance Table
##
## Response: F1$NBI
## Df Sum Sq Mean Sq F value Pr(>F)
## F1$Row 49 7866 160.527 2.3456 6.383e-07 ***
## F1$Column 3 789 263.021 3.8432 0.00934 **
## F1$Row:F1$Column 106 7724 72.864 1.0647 0.31303
## Residuals 1633 111761 68.439
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1FlavDoc<-read.xlsx("Flav.xlsx", sheetName = "Flav")
heatmaply(FlavDoc, xlab = "Column",ylab ="Row", Rowv = FALSE, Colv = FALSE)FRC <-lm(F1$Flav ~ F1$Row + F1$Column)
anova(FRC)## Analysis of Variance Table
##
## Response: F1$Flav
## Df Sum Sq Mean Sq F value Pr(>F)
## F1$Row 49 8.215 0.16765 2.923 1.283e-10 ***
## F1$Column 3 2.194 0.73147 12.754 3.031e-08 ***
## Residuals 1739 99.738 0.05735
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1FRC2 <- lm(F1$Flav ~ F1$Row*F1$Column)
anova(FRC2)## Analysis of Variance Table
##
## Response: F1$Flav
## Df Sum Sq Mean Sq F value Pr(>F)
## F1$Row 49 8.215 0.16765 3.2197 1.370e-12 ***
## F1$Column 3 2.194 0.73147 14.0479 4.826e-09 ***
## F1$Row:F1$Column 106 14.709 0.13876 2.6649 4.222e-16 ***
## Residuals 1633 85.030 0.05207
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1Block and Height
BHC <-lm(F1$Chl ~ F1$Row+F1$Column+F1$Height)
anova(BHC)## Analysis of Variance Table
##
## Response: F1$Chl
## Df Sum Sq Mean Sq F value Pr(>F)
## F1$Row 49 21394 436.61 1.9621 9.404e-05 ***
## F1$Column 3 2118 706.12 3.1733 0.02337 *
## F1$Height 2 730 365.17 1.6410 0.19408
## Residuals 1737 386522 222.52
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1BHF<-lm(F1$Flav ~ F1$Row+F1$Column+F1$Height)
anova(BHF)## Analysis of Variance Table
##
## Response: F1$Flav
## Df Sum Sq Mean Sq F value Pr(>F)
## F1$Row 49 8.215 0.16765 2.9483 8.704e-11 ***
## F1$Column 3 2.194 0.73147 12.8639 2.589e-08 ***
## F1$Height 2 0.969 0.48448 8.5202 0.0002079 ***
## Residuals 1737 98.769 0.05686
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1BHA<-lm(F1$Anth ~ F1$Row+F1$Column+F1$Height)
anova(BHA)## Analysis of Variance Table
##
## Response: F1$Anth
## Df Sum Sq Mean Sq F value Pr(>F)
## F1$Row 49 0.2346 0.0047878 2.5748 2.363e-08 ***
## F1$Column 3 0.0095 0.0031656 1.7024 0.1645
## F1$Height 2 0.0520 0.0260055 13.9853 9.433e-07 ***
## Residuals 1737 3.2299 0.0018595
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1BHN<-lm(F1$NBI ~ F1$Row+F1$Column+F1$Height)
anova(BHN)## Analysis of Variance Table
##
## Response: F1$NBI
## Df Sum Sq Mean Sq F value Pr(>F)
## F1$Row 49 7866 160.527 2.3352 7.084e-07 ***
## F1$Column 3 789 263.021 3.8262 0.009551 **
## F1$Height 2 80 39.808 0.5791 0.560512
## Residuals 1737 119404 68.742
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1Compare Parents
Parent <- rbind(EC201,EC103)
t.test(Parent$Chl ~ Parent$Tree.ID)##
## Welch Two Sample t-test
##
## data: Parent$Chl by Parent$Tree.ID
## t = -0.003842, df = 17.804, p-value = 0.997
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## -7.830954 7.802387
## sample estimates:
## mean in group EC103 mean in group EC201
## 52.59242 52.60670P1 <- lm(Parent$Chl ~ Parent$Tree.ID)
anova(P1)## Analysis of Variance Table
##
## Response: Parent$Chl
## Df Sum Sq Mean Sq F value Pr(>F)
## Parent$Tree.ID 1 0.0 0.001 0 0.9969
## Residuals 20 1461.7 73.087t.test(Parent$Flav ~ Parent$Tree.ID)##
## Welch Two Sample t-test
##
## data: Parent$Flav by Parent$Tree.ID
## t = -1.141, df = 19.612, p-value = 0.2676
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## -0.31409194 0.09215861
## sample estimates:
## mean in group EC103 mean in group EC201
## 1.599833 1.710800P2 <-lm(Parent$Flav ~ Parent$Tree.ID)
anova(P2)## Analysis of Variance Table
##
## Response: Parent$Flav
## Df Sum Sq Mean Sq F value Pr(>F)
## Parent$Tree.ID 1 0.06717 0.067165 1.2892 0.2696
## Residuals 20 1.04195 0.052098t.test(Parent$NBI ~ Parent$Tree.ID)##
## Welch Two Sample t-test
##
## data: Parent$NBI by Parent$Tree.ID
## t = 0.47573, df = 14.7, p-value = 0.6413
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## -5.076667 7.987334
## sample estimates:
## mean in group EC103 mean in group EC201
## 33.14333 31.68800P3<- lm(Parent$NBI ~ Parent$Tree.ID)
anova(P3)## Analysis of Variance Table
##
## Response: Parent$NBI
## Df Sum Sq Mean Sq F value Pr(>F)
## Parent$Tree.ID 1 11.55 11.553 0.2455 0.6256
## Residuals 20 941.01 47.050t.test(Parent$NBI ~ Parent$Tree.ID)##
## Welch Two Sample t-test
##
## data: Parent$NBI by Parent$Tree.ID
## t = 0.47573, df = 14.7, p-value = 0.6413
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## -5.076667 7.987334
## sample estimates:
## mean in group EC103 mean in group EC201
## 33.14333 31.68800P4 <-lm(Parent$NBI ~ Parent$Tree.ID)
anova(P4)## Analysis of Variance Table
##
## Response: Parent$NBI
## Df Sum Sq Mean Sq F value Pr(>F)
## Parent$Tree.ID 1 11.55 11.553 0.2455 0.6256
## Residuals 20 941.01 47.050Flowering
plot35<- ggplot(F1) + aes(x = Flower, y = Chl) + geom_boxplot()
plot36<- ggplot(F1) + aes(x = Flower, y = NBI) + geom_boxplot()
plot37<- ggplot(F1) + aes(x = Flower, y = Flav) + geom_boxplot()
plot38<- ggplot(F1) + aes(x = Flower, y = Anth) + geom_boxplot()
grid.arrange(plot35,plot36,plot37,plot38)
CF<- lm(F1$Chl ~ F1$Flower)
anova(CF)## Analysis of Variance Table
##
## Response: F1$Chl
## Df Sum Sq Mean Sq F value Pr(>F)
## F1$Flower 1 71 71.075 0.3098 0.5779
## Residuals 1790 410693 229.437NF<- lm(F1$NBI ~ F1$Flower)
anova(NF)## Analysis of Variance Table
##
## Response: F1$NBI
## Df Sum Sq Mean Sq F value Pr(>F)
## F1$Flower 1 196 196.270 2.7459 0.09768 .
## Residuals 1790 127943 71.476
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1FF<- lm(F1$Flav ~ F1$Flower)
anova(FF)## Analysis of Variance Table
##
## Response: F1$Flav
## Df Sum Sq Mean Sq F value Pr(>F)
## F1$Flower 1 0.677 0.67660 11.063 0.0008983 ***
## Residuals 1790 109.471 0.06116
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1AF<- lm(F1$Anth ~ F1$Flower)
anova(AF)## Analysis of Variance Table
##
## Response: F1$Anth
## Df Sum Sq Mean Sq F value Pr(>F)
## F1$Flower 1 0.0002 0.0001517 0.077 0.7814
## Residuals 1790 3.5259 0.0019698